ABSTRACT
Following the development of the simple US model of population and employment change, it seemed interesting to attempt to improve the representation of the system to describe the changes in employment in terms of the scheme of standard industrial indexes, with which economists and political administrations are familiar. Instead of crudely splitting economic activity into base and tertiary sectors, we want to deal with more familiar categories such as manufacturing industry, finance etc. In addition, instead of simply driving the model by reading values for the base employment, we want to allow the decision making processes within the model to drive the system as much as possible. This ‘causal mechanism’ approach, linked to the functional requirements of actors, should be contrasted with the alternative modelling philosophy based on the calibration of flows from data. In this chapter we develop a model of the changing pattern of economic activity and residential populations in the 9 Belgian Provinces. (Sanglier and Allen, 1989)
Our model is based on the dynamic spatial interaction of the different urban actors governing the locations and numbers of jobs and residents. Each variable represents the spatial behaviour of a typical actor of that category as a function of his locational preferences. As the model is based on the behaviour of the actors, it can be applied to several levels of the ‘urban structure’. In this application the spatial unit is the Province. The following scheme (Figure 6.1) shows the interactions and the flows between each province and the others, which lead to the migration of economic activities and people between provinces, and thus induce long-term structural changes.
